Towards a Dynamic Spatial Microsimulation Model for

The main advantages of behavioural MSMs are the ability to account for the heterogeneity within the population of interest, and the identification of both the mean and the distributional impact of a reform. [...] Then, comparing the values of the random variable and the predicted probabilities, the model determines whether the person is a mover or not in the current year. [...] We expect that if the simulated proportion of people changing their location, the proportion of people in each ethnic group, and the proportion of each ethnic group changing their location are close to the actual proportions, then the model should be able to replicate the actual levels of ethnic diversity and residential sorting in the Auckland population in 2006. [...] Step 2: Calibrating the proportion of people in each ethnic group We calculated the difference between the proportion of people in each ethnic group between the simulated data and the actual data. [...] The inconsistencies in the ethnic categorisations in the 1996 and 2001 Census data mentioned in Section 3, which were used to parameterise the initial model, contribute to the model performance.